Implementasi Particle Swarm Optimization untuk Optimasi Fuzzy-Social Force Model pada Sistem Navigasi Robot Omnidirectional

Abstract

Particle Swarm Optimization (PSO) is a swarm-based optimization method that is easy to implement and requires only a few parameters to set. This study aims to implement PSO to optimize the Fuzzy-Social Force Model (FSFM). FSFM combines the Social Force Model (SFM) as a navigation algorithm and the Fuzzy Inference Rule (FIS) to produce adaptive gain on SFM to create a mobile robot navigation system that is more responsive to obstacles. The PSO implementation optimizes fuzzy rules to be more optimal when the mobile robot navigates into social spaces. From the experimental test results on the VREP simulation software, cognitive parameter c1 = 1 and social parameter c2 = 2 produced the best navigation performance compared to other test parameter values.